Ethical Demand
It’s Time for Data Sharing by Pharmaceutical Companies
The data gathered by drug companies doing clinical trials is considered proprietary by many corporate managers, who point to the great expense a company incurs in going through the whole drug development and approval process. Still, an ethical demand exists for sharing data, in the public interest and in the service of curing disease.
In an examination of how well drug companies share clinical trial results and information about actual patients, Ben Goldacre and colleagues, writing in Fierce Biotech (www.bmj.com), found that 96 percent of the 23 biggest companies do havepolicies that allowed for synopses of both trial results and individual patient data (IPD) to be shared. Also, 74 percent allowed some disclosure of data from past trials, which often involve drugs now actively in use.
Study participant Sile Lane said, “Lack of policy on past trials is the biggest issue. Every day, as people and software retire, these are being lost. These include the trials on the treatments we’re using today, so they matter for patients now but also because they are the basis for measuring the effectiveness of proposed new treatments. The best evidence we have from academic research suggests that we can’t see the results from around half of these trials.”
The study found that while there is much encouraging news, there is also a widespread failure to include firm dates for sharing, and in some cases unclear language is used in reporting results.
The transparency study looked at the public or intra-industry release of two types of data: clinical study reports (CSRs) and individual patient data (IPDs).
CSRs are massive reports required by the Food and Drug Administration (FDA) but often overlooked by academic reviewers. Goldacre and colleagues cited a study that showed CSRs having double the information about a new drug’s risks and benefits than is found in other sources.
IPDs are the full information about each subject gathered during a trial, which allows outside researchers to review the whole process and draw their own conclusions.
Along with the volume of material and the timeliness of release, the study also found problems with the language some companies used in their transparency policy statements, such as the use of “considered for release” or release of “many” of the results. Different press releases from a company sometimes had inconsistent or contradictory statements about transparency.
After discovering the uneven performance of the drug industry on the critical issue of data sharing, the study proposed a “Model Clinical Trial Transparency Policy” with four points:
1. Registration: public announcement of all clinical trials before they begin.
2. Summary methods and results: issued for all trials within a stated period of time and available for free or on either clinicaltrials.gov or the company’s website.
3. CSR, shared completely for all trials within a stated time.
4. IPDs, to be shared on request within a stated time.
While pharma is attempting to do better at clinical trial transparency, more needs to be done. Realistic objectives and timelines can make that happen.
